Pulling Out All the Tops with Computer Vision and Deep Learning

Abstract

We apply computer vision with deep learning -- in the form of a convolutional neural network (CNN) -- to build a highly effective boosted top tagger. Previous work (the "DeepTop" tagger of Kasieczka et al) has shown that a CNN-based top tagger can achieve comparable performance to state-of-the-art conventional top taggers based on high-level inputs. Here, we introduce a number of improvements to the DeepTop tagger, including architecture, training, image preprocessing, sample size and color pixels. Our final CNN top tagger outperforms BDTs based on high-level inputs by a factor of $\sim 2$--3 or more in background rejection, over a wide range of tagging efficiencies and fiducial jet selections. As reference points, we achieve a QCD background rejection factor of 500 (60) at 50\% top tagging efficiency for fully-merged (non-merged) top jets with $p_T$ in the 800--900 GeV (350--450 GeV) range. Our CNN can also be straightforwardly extended to the classification of other types of jets, and the lessons learned here may be useful to others designing their own deep NNs for LHC applications.

33 pages, 11 figures

References in corpus (28)

• An Introduction to PYTHIA 8.2
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Herwig++ Physics and Manual
• Herwig 7.0 / Herwig++ 3.0 Release Note
• Jet substructure as a new Higgs search channel at the LHC
• Identifying Boosted Objects with N-subjettiness
• Soft Drop
• Top-tagging: A Method for Identifying Boosted Hadronic Tops
• Towards Jetography
• CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks
• Fat Jets for a Light Higgs
• Classification without labels: Learning from mixed samples in high energy physics
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Stop Reconstruction with Tagged Tops
• Deep learning in color: towards automated quark/gluon jet discrimination
• Jet-Images -- Deep Learning Edition
• Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multi-Layer Calorimeters
• Deep-learning Top Taggers or The End of QCD?
• Jet Substructure Classification in High-Energy Physics with Deep Neural Networks
• Deep-learned Top Tagging with a Lorentz Layer
• Jet-Images: Computer Vision Inspired Techniques for Jet Tagging
• Finding top quarks with shower deconstruction
• Finding physics signals with shower deconstruction
• Top Tagging
• Learning to Classify from Impure Samples with High-Dimensional Data
• Weakly Supervised Classification in High Energy Physics
• (Machine) Learning to Do More with Less

Cited by in corpus (95)

• ParticleNet: Jet Tagging via Particle Clouds
• Searching for New Physics with Deep Autoencoders
• Energy Flow Networks: Deep Sets for Particle Jets
• The Machine Learning Landscape of Top Taggers
• Learning New Physics from a Machine
• Extending the Bump Hunt with Machine Learning
• Graph Neural Networks in Particle Physics
• QCD or What?
• Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques
• Novelty Detection Meets Collider Physics
• Adversarially-trained autoencoders for robust unsupervised new physics searches
• JEDI-net: a jet identification algorithm based on interaction networks
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• Uncovering latent jet substructure
• Performance of top-quark and $W$-boson tagging with ATLAS in Run 2 of the LHC
• Autoencoders for unsupervised anomaly detection in high energy physics
• Tag N' Train: A Technique to Train Improved Classifiers on Unlabeled Data
• CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows
• DisCo Fever: Robust Networks Through Distance Correlation
• Machine Learning Uncertainties with Adversarial Neural Networks
• Interaction networks for the identification of boosted $H\to b\overline{b}$ decays
• ABCDisCo: Automating the ABCD Method with Machine Learning
• Quark-Gluon Tagging: Machine Learning vs Detector
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Deep-Learning Jets with Uncertainties and More
• Casting a graph net to catch dark showers
• Challenges for Unsupervised Anomaly Detection in Particle Physics
• Mapping Machine-Learned Physics into a Human-Readable Space
• Supervised Jet Clustering with Graph Neural Networks for Lorentz Boosted Bosons
• PC-JeDi: Diffusion for Particle Cloud Generation in High Energy Physics
• Boosted $W/Z$ Tagging with Jet Charge and Deep Learning
• Symmetries, Safety, and Self-Supervision
• Mass Unspecific Supervised Tagging (MUST) for boosted jets
• Equivariant Energy Flow Networks for Jet Tagging
• Reweighting a parton shower using a neural network: the final-state case
• Per-Object Systematics using Deep-Learned Calibration
• Measuring QCD Splittings with Invertible Networks
• Automating the Construction of Jet Observables with Machine Learning
• Topological Obstructions to Autoencoding
• Exploring anomalous couplings in Higgs boson pair production through shape analysis
• The forgotten channels: charged Higgs boson decays to a $W^{\pm }$ and a non-SM-like Higgs boson
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• Mass Agnostic Jet Taggers
• Infrared Safety of a Neural-Net Top Tagging Algorithm
• Binary JUNIPR: an interpretable probabilistic model for discrimination
• Jet Flavour Tagging for Future Colliders with Fast Simulation
• Neural Network-based Top Tagger with Two-Point Energy Correlations and Geometry of Soft Emissions
• Detecting an axion-like particle with machine learning at the LHC
• Higgs boson tagging with the Lund jet plane
• Reports of My Demise Are Greatly Exaggerated: $N$-subjettiness Taggers Take On Jet Images
• OmniJet-$\alpha$: The first cross-task foundation model for particle physics
• What's Anomalous in LHC Jets?
• Machine Learning in High Energy Physics: A review of heavy-flavor jet tagging at the LHC
• Topology classification with deep learning to improve real-time event selection at the LHC
• Model-agnostic search for dijet resonances with anomalous jet substructure in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• Combine and Conquer: Event Reconstruction with Bayesian Ensemble Neural Networks
• CapsNets Continuing the Convolutional Quest
• Six Top Messages of New Physics at the LHC
• Disentangling Boosted Higgs Boson Production Modes with Machine Learning
• Jet tagging algorithm of graph network with HaarPooling message passing
• Leveraging universality of jet taggers through transfer learning
• Reconstructing boosted Higgs jets from event image segmentation
• Invisible Higgs search through Vector Boson Fusion: A deep learning approach
• A Detailed Study of Interpretability of Deep Neural Network based Top Taggers
• Anomaly Awareness
• Searches for new physics with boosted top quarks in the MadAnalysis 5 and Rivet frameworks
• Jet Tagging with More-Interaction Particle Transformer
• Detector signal characterization with a Bayesian network in XENONnT
• Identifying quenched jets in heavy ion collisions with machine learning
• Maximum performance of strange-jet tagging at hadron colliders
• Quark jet rates and quark/gluon discrimination in multi-jet final states
• Towards recognizing the light facet of the Higgs Boson
• BIP: Boost Invariant Polynomials for Efficient Jet Tagging
• Using Machine Learning to Improve Neutron Identification in Water Cherenkov Detectors
• Pileup and Infrared Radiation Annihilation (PIRANHA): A Paradigm for Continuous Jet Grooming
• Accuracy versus precision in boosted top tagging with the ATLAS detector
• Deep Learning Jet Substructure from Two-Particle Correlation
• Boosted top tagging and its interpretation using Shapley values
• Jet Classification Using High-Level Features from Anatomy of Top Jets
• Automatic detection of boosted Higgs boson and top quark jets in an event image
• Unveiling the Secrets of New Physics Through Top Quark Tagging
• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• Jet Single Shot Detection
• Investigating top tagging with Y${}_{\text{m}}$-Splitter and N-subjettiness
• Search for new resonances decaying to pairs of merged diphotons in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• Probing sub-TeV Higgsinos aided by a ML-based top tagger in the context of Trilinear RPV SUSY
• Top-philic Machine Learning
• Jet Reconstruction with Mamba Networks in Collider Events
• Deep Image Reconstruction for Background Subtraction in Heavy-Ion Collisions
• Search for resonances decaying to an anomalous jet and a Higgs boson in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• The Physics Behind ML-based Quark-Gluon Taggers
• Tagging ultra-boosted jets at FCC-hh using machine learning techniques
• Theory, phenomenology, and experimental avenues for dark showers: a Snowmass 2021 report
• Creating Simple, Interpretable Anomaly Detectors for New Physics in Jet Substructure
• Pulling the Higgs and Top needles from the jet stack with Feature Extended Supervised Tagging